Numerical modelling and inverse analysis of continuous compaction control

Abstract

Abstract Assessing the dynamic properties of compacted materials placed in field remains a challenge. Currently, stiffness parameters of compacted fills are estimated by means of spot tests such as dry density measurement, load-plate or CBR field tests, and more recently, Continuous Compaction Control (CCC). Unfortunately, these methodologies are based on non-constitutive parameters and correlations. In this work, CCC is considered a dynamic field test that is useful to evaluate properties commonly used in earthworks design, namely: Young modulus, hysteretic damping, and Poisson’s ratio. Considering CCC load-displacement histories as an output of this field test, a parametric study within the framework of continuum mechanics could be performed to fit the CCC data test. However, a parametric study to solve such an inverse problem is time-consuming; and, moreover, it does not guarantee a proper solution. Therefore, this paper proposes a data reduction methodology that generates a solution of this inverse problem by using at least six numerical simulations within continuum mechanics. A Finite Differences code was used with a simplified version of Kelvin-Voigt viscoelastic model to solve the wave equation. In numerical simulations, field drum-soil load histories, extracted from literature, were applied to a 2D numerical soil model. Data from three CCC field tests were employed to test the usefulness and validity of the proposed data reduction methodology. Validations show its usefulness not only for full contact drum-soil case but also for the partial loss of contact. Although dynamic numerical simulations are time-consuming and depend on computer resources, real-time application is promising. This attempt of the evaluation of mechanical properties of compacted soils from CCC records shows the potential use of CCC as a dynamic field test for analysis and design of earthworks such as embankments and pavements.